Post on 18-Mar-2020
transcript
Targeted Screening and
Treatment of Latent Tuberculosis
Infection in Hong Kong
Kwok-Chiu CHANG
TB & Chest Service
Public Health Services Branch
Centre for Health Protection
Department of Health, Hong Kong SAR
3 May 2016
Auramine-O staining of AFB under Fluorescence Microscopy
Natural history of tuberculosis
Exposure Infection Disease
TB infection
No disease, but
latent infection with
TST +ve : 95%
Primary TB :
5% (5 yr)
Infection remaining
latent for life : 90%
Reactivation and
developing disease ~ 5%
life-time risk
Tuberculosis (TB) notification rates Year TB notification TB
deaths
Death /
Notifications (%)
TB notification rate
(per 100,000 population)
Total Chinese immigrants
< 7 years
2010 5093 80 191 3.75 72.5
2011 4794 81 187 3.90 67.8
2012 4858 100 199 4.10 67.9
2013 4664 92 178 3.82 64.9
2014 4705 85 187 3.97 65.0
2015* 4498 81 167 3.71 61.6
* Provisional Figure
Ref: Annual Report 2013. TB & Chest Service, Department of Health, Hong Kong
1950
TB
cases
2012
Ageing of the TB epidemic
Endogenous reactivation now accounts for majority
of TB cases in most Intermediate Burden Areas
Epidemiological Transition
TB notification rates by age groups Age Group TB notification rate (per 100,000 population) in 2015 *
< 10 < 2
10-14 6.82
15-19 28.48
20-49 39.83-45.18
50-54 51.33
55-59 71.77
60-64 87.06
65-69 100.47
70-74 138.06
75-79 159.62
80-84 212.75
≥ 85 277.35
Ref: http://www.chp.gov.hk/en/data/1/10/26/43/5104.html
* Provisional Figure
What is latent infection with M.
tuberculosis (LTBI)?
LTBI is a spectrum Clinical disease
Subclinically active infection: Bacterial replication
maintained at a subclinical level by the immune system
Quiescent infection: Infection controlled with some bacteria
persisting in a non-replicating form
Infection eliminated by acquired immunity in associated
with T cell priming
Infection eliminated by innate immunity without priming
antigen-specific T cells Reference: Barry et al. Nat Rev Microbiol 2009; 7: 845-855
Diagnostic tools for LTBI Measurement of specific host immune
responses
Methods
Tuberculin skin test (traditional standard)
Use PPD RT 23 two units
Interferon-γ release assay (IGRA)
More specific antigens: ESAT6, CFP10, (TB 7.7)
T-Spot.TB® (Oxford Immunotec) (Separate monocyte layer from fresh blood)
QuantiFERON® -TB Gold / In-tube (Cellestis) (Fresh whole blood)
Serological tests: not useful
Mantoux test: intradermal injection of
Tuberculin Purified Protein Derivative
Reference: Oxford Immunotec
ELISPOT =
enzyme-linked immunospot
Reference: Qiagen
Blood sampling and incubation
IFN-gamma ELISA
ELISA =
enzyme-linked
immunosorbent
assay
Latent infection with M. tuberculosis
Pragmatically defined as presumptive infection
with M. tuberculosis complex, as evidenced by
a positive tuberculin skin test (TST) reaction
and/or a positive interferon-ϒ release assay
(IGRA) result without any sign of clinically or
radiologically manifest disease.
Asymptomatic and NON-INFECTIOUS
Do not notify LTBI.
TST vs. IGRA: differences (1) TST QuantiFERON TB (QFT)
Gold/ In-Tube (IT)
T-SPOT.TB
Nature In vivo Ex vivo Ex vivo
Antigens Mixture in PPD;
shared by BCG
More TB-specific: ESAT-6,
CFP-10, (TB7.7)
More TB-specific:
ESAT-6, CFP-10
Interference by
BCG
Yes No No
Exposure
correlation
Some, especially if
not BCG-vaccinated
Higher Higher/ highest?
Method Manually measuring
induration, read 48-
72 hours later
ELISA (can be automated),
plasma can be stored for ~
8 weeks at 4 o C
ELISPOT (can be
automated), completed
within 2 days
Need of support
by laboratory
No High; fresh blood delivery Highest; fresh blood
and cell separation
Inter-reader
variability
Can be substantial Minimal Minimal
No. of visits 2 1 1
Reference: Leung CC et al. Eur Respir J. 2011;37:690-711
ELISA = enzyme-linked immunosorbent assay ELISPOT = enzyme-linked immunospot
TST vs. IGRA: differences (2) TST QFT TB Gold/ IT T-SPOT.TB
Choice of cutoff 5, 10, 15 mm by clinical
scenario; higher disease
risk with larger induration
Single Single
Sensitivity
specificity trade-off
Yes Not fully clarified Not fully clarified
Advanced age Significantly affected Less affected Less affected
Immune
compromise
Significantly affected Less affected Less affected/
Least?
Proxy sensitivity 71-82% QFT-Gold: 73-82%
QFT-Gold IT:63-78%
86-93%
Proxy specificity No BCG: 95-99%
BCG: low, heterogeneous
No BCG: 98-100%
BCG: 94-98%
86-100%
Conversion Criteria established Not fully clarified Not fully clarified
Booster effect Yes No (prior TST may
affect)
No (prior TST
may affect)
Longitudinal data Abundant Less Less
Reference: Leung CC et al. Eur Respir J. 2011;37:690-711
TST vs. IGRA: similarities
Both are indirect measures of TB infection
NEITHER rules in active TB
NEITHER rules out
active TB or LTBI in the immunocompromised
NEITHER tells
recent vs. remote infection
NEITHER determines
treatment efficacy
Species Specificity of IGRA TB Antigens
Species ESAT-6 CFP-10 TB7.7
M. tuberculosis + + +
M. africanum + + +
M. bovis + + +
BCG strains - - -
Most NTM - - -
M. kansasii + + +
M. marinum + + +
M. szulgai + + +
ESAT-6 = 6-kDa early secretory antigenic target
CFP-10 = 10-kDa culture filtrate protein
Ref: Leung CC, et al. Eur Respir J. 2011;37(3):690-711
Screening for LTBI Universal vs. Targeted
Targeted screening is the
currently recommended approach.
Ref: Leung CC, et al. Eur Respir J. 2011;37(3):690-711
Targeted screening for LTBI
Target = high-risk groups
Screening = intention to treat in order
to prevent TB disease
Number needed to treat (NNT) to prevent 1
TB case in 5 years
0
50
100
150
200
250
300
350
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0
Num
ber
need
ed to
tre
at t
o pr
even
t 1
TB
cas
e
Cumulative incidence in 5 years, %
RR=0.4
RR=0.25
RR=0.1
Reference: Leung CC et al. Eur Respir J. 2011;37:690-711
Cumulative incidence in 5 years (%)
NNT to
prevent
1 TB
case
Benefit versus Risk Ratio (Number of TB
cases prevented per case of hepatitis)
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
0.25
0.50
0.75
1.00
1.25
1.50
1.75
2.00
2.25
2.50
2.75
3.00
3.25
3.50
3.75
4.00
4.25
4.50
4.75
5.00
Cumulative incidence in 5 years, %
Ben
efit
vs
Ris
k R
atio
( T
B P
reve
nted
vs
Hep
atit
is)
Hepatitis Rate 0.4%
Hepatitis Rate 1.0%
Hepatitis Rate 2.0%
Hepatitis Rate 3.0%
Reference: Leung CC et al. Eur Respir J. 2011;37:690-711
Cumulative incidence in 5 years (%)
Number of
TB cases
prevented
per case of
drug-
related
hepatitis
Estimated Infection Rate (HK)
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0 - 4 5 - 9 10 -
14
15 -
19
20 -
24
25 -
29
30 -
34
35 -
39
40 -
44
45 -
49
50 -
54
55 -
59
60 -
64
65 -
69
70 -
74
75+
Age Groups
Pe
rce
nta
ge
*Estimation based on: Incidence (smear-positive cases) = ARI * Styblo ratio
Ref: Leung CC, et al. Eur Respir J. 2011;37(3):690-711
Who are the TB high-risk
groups that we may target?
Ref: Leung CC, et al. Eur Respir J. 2011;37(3):690-711
Incidence of disease among tuberculin-
positive subjects (per 1000 person-years
Relative Disease Risk
Recent TB infection Infection < 1 year past 12.9
Infection 1-7 year past
1.6
Old TB scar 2.0-13.6
HIV infection 35.0-162
Body Mass Index < 18.5 2.11
Diabetes mellitus (DM) DM vs no DM 1.8-4.1
HbA1c ≥ 7% vs. < 7% 3.1
Chronic renal failure 10.0-25.3
Renal Transplant 37
Heart Transplant 20-74
Head and neck carcinoma 16
Silicosis 68 30
Smoking Current smokers 2.63
Ex-smokers 1.41
Never smokers 1.00
Passive smoking 1.49
Incidence/ relative risk of active TB
TB Risk in RA (2004-08 HK cohort)
*Expected number of TB cases at the sex and age-adjusted rate of the Hong Kong population.
# Size of hypothetical control population of same sex and age mix as the RA cohorts and with similar sex and
age-adjusted TB rates as the Hong Kong population that would be expected to give the same number of
observed TB cases within the same period.
Independent explanatory variables associated with an increase risk of active TB included older age at study
entry (RR 1.05, p=0.013) a past history of pulmonary TB (RR 5.48, p=0.001), extra-pulmonary TB (RR 16.45,
p<0.001), Felty’s syndrome (RR 43.84, p=0.005), prednisolone >10mg daily (RR 4.44, p=0.009) and the use of
TNF blockers (RR 12.48, p<0.001).
RA cohorts No of
patients at
risk
Observed
TB cases
*Expected
TB cases
#Population
denominator
SIR 95% CI p-value
RA combined 2441 20 6.954 6972 2.876 1.55-5.35 <0.001 TNF blockers naïve RA 2424 16 6.839 5706 2.354 1.17-4.67 0.013 TNF blockers treated RA 81 4 0.115 2829 34.922 8.89-137.20 <0.001
~15x RA
Reference: Tam LS, et al. Clin Exp Rheumatol 2010;28:679-85.
Ref: TB & Chest Service, Department of Health of Hong Kong SAR. Guidelines on targeted tuberculin
testing and treatment of latent tuberculosis infection (Internal guidelines). 2015.
Targeted screening and
treatment of LTBI
Target group TST cutoff
1. Household contacts (esp. < 35
years old) of a smear+ source
Age > 1 year: 15 mm
Age < 1 year: 5 mm
2. Silicosis 10 mm
3. HIV-infected 5 mm
4. Immunosuppression/ Use of
biologics especially TNF blocker
Before immunosuppression: 10 mm
After immunosuppression: 5 mm*
* if not screened at baseline, or > baseline when screened again for a new indication
How good are TST/ IGRA for
predicting TB disease?
Which test to use?
Field performance of a test: factors Test characteristics
Sensitivity and specificity
Predictive values
Positive Predictive Value (PPV)
Negative Predictive Value (NPV)
Sensitivity, specificity, prevalence
Risk ratio (RR) or Incidence rate ratio (IRR)
RR = TB disease risk among test-positive/ TB
disease risk among test-negative
IRR: censoring, follow-up duration
RR = PPV/(1-NPV)
Systematic review: IGRA vs. TST
Test Pooled unadjusted IRR (95% CI)
IGRA 2.11 (1.29–3.46)
TST (cutoff = 10 mm) 1.60 (0.94–2.72)
TST (cutoff = 5 mm) 1.43 (0.75–2.72)
Ref: Rangaka MX et al. Lancet Infect Dis 2012; 12:45-55
Compared with test-negative results, IGRA-positive and TST-positive results
were much the same with regard to the risk of tuberculosis in five studies that
used both methods.
Until more predictive biomarkers are identified, existing tests for LTBI should be
chosen on the basis of relative specificity in different populations, logistics, cost,
and patients’ preferences rather than predictive ability alone.
Performance of IGRA could be better in high income areas, but potential bias
prevent firm conclusion.
IRR = incidence rate ratio
T-SPOT.TB vs. TST in predicting active TB
among patients with silicosis in Hong Kong T-SPOT.TB TST
(Cutoff = 5mm)
TST
(Cutoff = 10mm)
TST
(Cutoff = 15mm)
Test positive a 151 161 136 89
Test negative a 90 80 105 152
Active TB cases predicted b 12 9 9 4
Active TB cases missed b 1 4 4 9
Sensitivity % (95% CI) 92.3 (64.6-100.0) c 69.2 (42.0-87.7) 69.2 (42.0-87.7) 30.8 (12.4-58.0)c
Specificity % (95% CI) 39.0 (32.9-45.5) c 33.3 (27.5-39.7) 44.3 (38.0-50.8) 62.7 (56.3-68.7)c
PPV % (95% CI) 7.9 (4.5-13.5) 5.6 (2.8-10.4) 6.7 (3.4-12.3) 4.5 (1.4-11.4)
Positive results per case
predicted
13 (8-23) 18 (10-36) 16 (9-30) 23 (9-72)
NPV % (95%CI) 98.9 (93.4-100.0) 95.0 (87.5-98.4) 96.2 (90.3-98.8) 94.1 (89.0-97.0)
Negative results per case
missed
90 (16- >10000) 20 (8-63) 27 (11-84) 17 (10-34)
IRR (test positive/ test
negative)
8.50 (1.11-65.4)d 1.00 (0.31–3.25) 1.69 (0.52–5.50) 0.72 (0.22–2.33)d
Ref: Leung CC et al. Am J Respir Crit Care Med 2010;182:834-840
a. Excluding subjects treated for LTBI.
b. Prevalence of TB disease = 13/241 = 5.4%
c. T-SPOT.TB vs. TST (cutoff) with P < 0.001
d. T-SPOT.TB vs. TST (cutoff) with P < 0.05
T-SPOT.TB vs. TST in predicting active TB
among household contacts in Hong Kong T-SPOT.TB TST
(Cutoff = 5mm)
TST
(Cutoff = 10mm)
TST
(Cutoff = 15mm)
Test positive a 244 503 337 119
Test negative a 621 362 528 746
Active TB cases
predicted b
15 18 17 8
Active TB cases
missed b
5 2 3 12
Sensitivity % (95% CI) 75.0 (50.9–91.3) c 90.0 (68.3–98.8) 85.0 (62.1–96.8) 40.0 (19.1–63.9) c
Specificity % (95% CI) 72.9 (69.8–75.9) d 42.6 (39.2–46.0) d 62.1 (58.8–65.4) d 86.9 (84.4–89.1) d
PPV % (95% CI) 6.1(3.5–9.9) 3.6 (2.1–5.6) 5.0 (3.0–8.0) 6.7 (2.9–12.8)
Positive results per case
predicted (95% CI)
16 (10–29) 28 (18–47) 20 (13–34) 15 (8–34)
NPV % (95%CI) 99.2(98.1–99.9) 99.4 (98.0–99.9) 99.4 (98.3–99.9) 98.4 (97.2–99.5)
Negative results per
case missed (95% CI)
124 (54–382) 181 (50–1493) 176 (61–853) 62 (36–120)
IRR (test positive vs.
test negative)
7.7 (2.8–21.2) 6.4 (1.5–27.5) 8.8 (2.6–30.1) 4.2 (1.7–10.3)
Ref: Leung CC et al. Respirology 2015;20:496-503
a. Excluding subjects treated for LTBI.
b. Prevalence of TB disease = 20/865 = 2.3%
c. T-SPOT.TB vs. TST (cutoff) with P < 0.05
d. T-SPOT.TB vs. TST (cutoff) with P < 0.001
Targeted screening for LTBI among
the HIV-infected in Hong Kong
TST: cutoff at 5 mm Treatment is indicated for HIV infected patients
with significant recent exposure to an infectious source of TB regardless of TST results.
IGRA is an acceptable alternative.
Dual testing by TST and IGRA: Advisable when CD4 count <100/μL
A positive result with either test is an indication for treatment.
Blood should be drawn for IGRA before or on the same day as placing the TST to avoid potential PPD sensitisation.
Ref: Scientific Committee on AIDS and STI (SCAS), Centre for Health Protection, Department of Health.
Recommendations on the Management of Human Immunodeficiency Virus and Tuberculosis Coinfection. March 2015
Treatment of LTBI: assuming no drug resistance
in the source case
LTBI: Isoniazid (INH) Preventive Treatment
6-12 m: Efficacy Meta-analysis
73375 subjects, 11 RCT, HIV/non-HIV, 6-12 months
Risk reduction: 60% [Risk ratio 0.40 (95% CI 0.31-0.52)]
IUAT trial 1970s, Eastern Europe, 28000 subjects, previously
untreated fibrotic lesions
Risk reduction: 21% (3H), 65% (6H), 75% (12H)
Risk reduction (good compliance): 30% (3H), 69% (6H), 93% (12H)
Cost-effectiveness analysis: 6H best (adopted by most public health programmes)
Comstock (1999): 9-10H conferred optimal protection
Efficacy (9H) ~ 90% (ATS/ US CDC recommends) References: 1. Leung CC et al. Eur Respir J 2011; 37: 690–711
2. Chee CB et al. Respirology 2013;18:205–216
LTBI: Isoniazid (INH) Preventive Treatment
Treatment
regimen
Studies Efficacy
6H vs. 36H RCT
(Botswana) [Ref: Samandari T
et al Lancet.
2011;377:1588-98]
ART was provided to those with CD4 <200/mL.
36H was more effective than 6H for preventing TB in
those who were TST-positive.
ART independently reduced TN incidence by 50%.
3HP vs. 3HR
vs. continuous
H (up to 6
years) vs. 6H
RCT (South
Africa) [Ref: Martinson
NA et al N Engl J
Med 2011;365:11-
20]
Median CD4 484/mL
ART not given
None of the alternative regimens were superior to 6H.
Serious adverse reactions more common in the
continuous H group than in the other groups (18.4 per
100 person-years vs. 8.7–15.4 per 100 person-years).
Prolonged INH therapy Non-HIV infected: > 12m NOT useful
HIV-infected: CONFLICTING findings from RCTs
References: 1. Leung CC et al. Eur Respir J 2011; 37: 690–711
2. Chee CB et al. Respirology 2013;18:205–216
LTBI: Isoniazid (INH) Preventive Treatment Primary INH prophylaxis RCT: No improvement
HIV-infected children: TB disease-free survival
HIV-uninfected BCG-vaccinated children: TB infection-free survival
[Madhi SA et al. N Engl J Med. 2011;365:21-31]
References: 1. Leung CC et al. Eur Respir J 2011; 37: 690–711.
Risk of INH-related hepatitis
Age group Risk
20 - 34 0.3%
35 - 49 1.2%
50 - 64 2.3%
> 64 4.6%
Ref: Kopanoff DE, et al. Am Rev Respir Dis 1978;117:991.
LTBI: Alternative regimens (1) Rifampicin monotherapy (4 months): 4R
4 RCT (vs. 9H): well-tolerated (3), efficacy data (1)
Meta-analysis of 4 RCT
non-completion (relative risk 0.53; 95% CI
0.44–0.63)
hepatotoxicity (relative risk 0.12; 95% CI
0.05–0.30)
cost-effective, US$213 savings per patient
treated
A multicentre trial comparing 4R vs. 9INH
ongoing
References: 1. Leung CC et al. Eur Respir J 2011; 37: 690–711
2. Chee CB et al. Respirology 2013;18:205–216
LTBI: Alternative regimens (2)
Isoniazid + rifampicin (3 months): 3HR
Hong Kong Silicosis Study
RCT (6H vs. 3HR vs. 3R vs. placebo),
Risk ratio (95% CI): 6H = 0.58 (0.36-0.94) vs.
3HR = 0.64 (0.39-1.03) vs. 3R = 0.48 (0.29-0.80)
Efficacy ~ 36%
Meta-analysis of 5 RCT (3-4 months HR)
6H (3), 9H (1), 12H (1, HIV-infected)
3HR not different from 6-12 INH therapy by
efficacy, severe side-effects and mortality
Overall efficacy ~ 65% References: 1. Leung CC et al. Eur Respir J 2011; 37: 690–711
2. Chee CB et al. Respirology 2013;18:205–216
LTBI: Alternative regimens (3,4)
Isoniazid + rifapentine (weekly for 12 weeks): 3HP
RCT (Schechter M et al, 2006): largely HIV-uninfected
RCT (Martinson NA et al, 2011): HIV-infected
RCT (Sterling TA et al, 2011): largely HIV-uninfected
Rifampicin plus pyrazinamide (2 months) [NOT recommended]
excess hepatotoxicity in non-HIV infected
RCT: 3HP (Sterling TA et al, 2011)
US, Canada, Brazil, Spain
3HP vs. 9H
7731 subjects, largely HIV-uninfected, followed
for 33 months
Non-inferiority trial: non-inferiority margin=0.75%
Cumulative TB incidence 0.19% (3HP) and 0.43% (9H)
Difference = 0.24%
3HP is non-inferior to 9H
Effective in immunocompentent and HIV-infected
Rifamycin resistance: no evidence (too few isolates)
Treatment completion is more likely
with 3HP than 9H
Permanent drug
discontinuation
Isoniazid daily for
9 months
(n = 3759)
Isoniazid and Rifapentine
once weekly for 12 weeks
(n = 4040)
P value
For any reason 1160/3745 (31.0%) 713/3986 (17.9%) <0.001
Because of an
adverse event
139/3745 (3.7%) 196/3986 (4.9%) 0.009
Reference: Sterling TR et al. N Engl J Med 2011;365:2155-66
3HP is less hepatotoxic than 9H
Adverse events
attributable to
drug
Isoniazid daily
for 9 months
(n = 3759)
Isoniazid and Rifapentine
once weekly for 12 weeks
(n = 4040)
P value
Related to drug 206 (5.5%) 332 (8.2%) <0.001
Hepatotoxicity 103 (2.7%) 18 (0.4%) <0.001
Rash 21 (0.6%) 31 (0.8%) 0.26
Possible
hypersensitivity
17 (0.5%) 152 (3.8%) <0.001
Others 65 (1.7%) 131 (3.2%) <0.001
Reference: Sterling TR et al. N Engl J Med 2011;365:2155-66
Treatment of LTBI: assuming drug resistance in
the source case
Preventive treatment for
contacts of drug-resistant TB
Reference: Chee CB et al. Respirology 2013;18:205–216
Scenario Evidence Options
H-resistant
TB
contacts
Case series
Contacts given 6R did
not develop active TB
Contacts given isoniazid
monotherapy had the
same rate of active TB as
those not given treatment.
4R or 6R
2RZ (generally not recommended)
MDR-TB
contacts
Systematic review
It is not possible to
support or reject the use
of preventive treatment.
At least two drugs (combinations of Z, E and/or
FQ) for 6–12 months according to source case’s
DST results (US CDC)
Fluoroquinolone monotherapy
High-dose H and a FQ (ideally levofloxacin) for
≥ 6 months in children under 5 or HIV-infected.
Close observation for ≥ 2 years for contacts
who are otherwise healthy and do not have risk
factors for rapid disease progression and
dissemination (UK health authorities, WHO,
and European CDC)
Summary
Operational definition of LTBI
Targeted screening of LTBI
It is unrealistic to screen and treat all.
Target at high-risk groups, more for personal
protection than public health
Screening is with an intent to treat.
Diagnostic tools
TST vs. IGRA
Parameters for evaluating diagnostic tools
Treatment options
Thank you